The authors have declared that no competing interests exist.
Conceived and designed the experiments: BS HC SM. Performed the experiments: HC KH. Analyzed the data: BS HC ET. Wrote the paper: BS HC.
Adult cigarette smokers usually become dependent on cigarettes during adolescence. Despite recent advances in addiction genetics, little data delineates the genetic factors that account for the vulnerability of humans to smoke tobacco. We studied the operant nicotine self-administration (SA) behavior of six inbred strains of adolescent male rats (Fisher 344, Brown Norway, Dark Agouti, Spontaneous Hypertensive Rat, Wistar Kyoto and Lewis) and six selected F1 hybrids. All rats were trained to press a lever to obtain food starting on postnatal day (PN) 32, and then nicotine (0.03 mg/kg/infusion, i.v.) reinforcement was made available on PN41-42 (10 consecutive daily 2 h sessions). Of the 12 isogenic strains, Fisher rats self-administered the fewest nicotine infusions (1.45±0.36/d) during the last 3 d, while Lewis rats took the most nicotine (13.0±1.4/d). These strains sorted into high, intermediate and low self-administration groups in 2, 2, and 8 strains, respectively. The influence of heredity on nicotine SA (0.64) is similar to that reported for humans. Therefore, this panel of isogenic rat strains effectively models the overall impact of genetics on the vulnerability to acquire nicotine-reinforced behavior during adolescence. Separate groups of rats responded for food starting on PN41. The correlation between nicotine and food reward was not significant. Hence, the genetic control of the motivation to obtain nicotine is distinctly different from food reward, indicating the specificity of the underlying genetic mechanisms. Lastly, the behavior of F1 hybrids was not predicted from the additive behavior of the parental strains, indicating the impact of significant gene-gene interactions on the susceptibility to nicotine reward. Taken together, the behavioral characteristics of this model indicate its strong potential to identify specific genes mediating the human vulnerability to smoke cigarettes.
Tobacco use is the single most preventable cause of disease, disability, and death in the United States. Approximately 20% of US adults smoke cigarettes. Each year, approximately 443,000 premature deaths are attributable to smoking or exposure to second hand smoke. Among adult smokers, 85% began before age 21 and 68% prior to 18
Nicotine is the principal psychoactive ingredient of tobacco products. The effect of nicotine on motivated behavior (i.e., wanting and using cigarettes) is often modeled using operant self-administration (SA) procedures. This model pertains to a variety of species, such as mice, rats, primates and even human. Amongst these, the rat is the most widely studied, and a variety of nicotine-modulated behaviors have been demonstrated, such as dependence, withdrawal, extinction, and relapse
The overall impact of genetic and genomic differences on the vulnerability to smoke has been estimated at ∼ 0.5 in numerous heritability studies
Rodent models can unambiguously identify candidate genes because both genetic and environmental factors are controllable. The success of such a model requires evidence of strong phenotypic variation in a smoking-specific behavior. This can be accomplished by identifying a panel of inbred rodent strains that differ in their nicotine SA behavior. More than 500 strains of inbred rats have been described
For each rat strain, the numbers of active v. inactive lever presses during the entire 10 d of nicotine SA are shown in
Six inbred and six isogenic F1 hybrid strains were trained to press a lever for food on postnatal day 33. Nicotine SA started on postnatal day 41 or 42. Nicotine (30 µg/kg/infusion, i.v.) was delivered using a fixed-ratio 1 schedule. Each session lasted 2 h. A total of 10 daily sessions were conducted without interruption. Statistical analyses of the number of lever presses and nicotine infusions for the last 3 d were shown in
Nicotine | Food | |||||
Strain | Df | F | p | Df | F | p |
BN | (1,9) | 22.04 |
|
(1,8) | 168.50 |
|
DA | (1,7) | 2.00 | 0.200 | (1,7) | 86.85 |
|
F344 | (1,7) | 8.70 |
|
(1,10) | 44.92 |
|
LEW | (1,9) | 16.53 |
|
(1,8) | 100.20 |
|
SHR | (1,8) | 82.99 |
|
(1,6) | 164.90 |
|
WKY | (1,9) | 18.78 |
|
(1,8) | 46.36 |
|
FD F1 | (1,7) | 10.70 |
|
(1,9) | 96.36 |
|
FL F1 | (1,9) | 23.73 |
|
(1,9) | 50.45 |
|
FS F1 | (1,10) | 14.40 |
|
(1,8) | 29.34 |
|
LB F1 | (1,10) | 13.56 |
|
(1,8) | 117.50 |
|
LS F1 | (1,5) | 7.83 |
|
(1,7) | 120.50 |
|
WL F1 | (1,7) | 9.93 |
|
(1,9) | 120.40 |
|
Active v. inactive lever presses (numbers not shown) during the last 3 d of nicotine or food SA were analyzed using repeated measures ANOVA for each isogenic strain. The p values that achieved statistical significance (p<0.05) are highlighted in bold and italics.
The means for the number of active and inactive lever presses as well as nicotine infusions are shown for the last 3 d of SA. Strains are ordered by the number of nicotine infusions. Results for pair-wise comparisons are shown in
Of these strains, F344 self-administered the fewest nicotine per day (1.45±0.36 infusions), while LEW took the most (13.0±1.4) – an 8.9-fold difference. Pair-wise comparisons between strains in the number of nicotine infusions are listed in
BN | DA | F344 | FD | FL | FS | LEW | LB | LS | SHR | WKY | |
|
1.000 | ||||||||||
|
1.000 | 1.000 | |||||||||
|
1.000 | 1.000 | 1.000 | ||||||||
|
|
|
|
|
|||||||
|
0.989 | 0.996 | 0.934 | 0.999 |
|
||||||
|
|
|
|
|
|
|
|||||
|
1.000 | 1.000 | 0.996 | 1.000 |
|
1.000 |
|
||||
|
|
|
|
|
1.000 |
|
|
|
|||
|
|
|
|
|
0.917 |
|
0.516 |
|
0.960 | ||
|
0.375 | 0.508 | 0.236 | 0.598 | 0.310 | 0.972 |
|
0.793 | 0.562 |
|
|
|
0.999 | 1.000 | 0.987 | 1.000 |
|
1.000 |
|
1.000 |
|
|
0.906 |
The mean number of nicotine infusions taken by each strain during the last 3 d of SA was compared using a post-hoc Tukey HSD procedure. The p values are shown for all the pair-wise comparisons. Comparisons that achieved statistical significance (p<0.05) are in bold and italics.
For each rat strain, the numbers of active vs. inactive lever presses during the entire 10 days of food SA are shown in
Six inbred and six isogenic F1 hybrid strains were trained to press a lever for food on postnatal day 33–36. Food reward resumed on postnatal day 41 or 42. Food pellets were delivered using a fixed-ratio 1 schedule. Each session lasted 2 h. A total of 10 daily sessions were conducted without interruption. Statistical analyses of the number of lever presses and food rewards earned are shown in
The means for the number of active and inactive lever presses, as well as food reward earned, are shown for the last 3 d of SA. Strains are ordered by the number of food rewards earned. Results for pair-wise comparisons are shown in Table 3.
Pair-wise comparisons between strains in the number of food pellets obtained during the last 3 d are shown in Table 3. LB rats were not different from BN and F344; these are the low cohort. The high cohort consists of SHR, FS and LS. The remaining strains (LEW, DA, WKY, FD, FL and WL) comprise the median group.
BN | DA | F344 | FD | FL | FS | LEW | LB | LS | SHR | WKY | |
|
1.000 | ||||||||||
|
0.990 | 0.728 | |||||||||
|
0.271 | 0.833 |
|
||||||||
|
0.081 | 0.501 |
|
1.000 | |||||||
|
|
|
|
|
|
||||||
|
1.000 | 1.000 | 0.925 | 0.496 | 0.193 |
|
|||||
|
0.064 |
|
0.520 |
|
|
|
|
||||
|
|
|
|
|
|
0.999 |
|
|
|||
|
|
|
|
|
|
0.996 |
|
|
1.000 | ||
|
0.330 | 0.870 |
|
1.000 | 1.000 |
|
0.564 |
|
|
|
|
|
|
|
|
0.702 | 0.951 | 0.604 |
|
|
0.138 | 0.106 | 0.711 |
The mean number of food pellets earned by each strain during the last 3 d of SA was compared using a post-hoc Tukey HSD procedure. The p values are shown for all the pair-wise comparisons. Comparisons that achieved statistical significance (p<0.05) are in bold and italics.
The mean number of active lever presses and rewards earned during the last 3 d by rats self-administering nicotine vs. obtaining food reward were compared across all strains (
The numbers of nicotine and food rewards earned by the 12 strains during the last 3 days were not statistically significant. The correlation of active lever presses during nicotine v. food SA was not significant.
The narrow-sense heritability for nicotine SA, calculated based on the mean nicotine intake during the last 3 d of SA, was 0.64, while the heritability for food reward was 0.71. The difference between F1 and the expected means of the two parental strains are listed in
Strain | Nicotine intake | Food Reward |
LB v. LEW & BN |
|
|
FD v. F344 & DA | 0.794 |
|
FL v. F344 & LEW |
|
|
WL v. WKY & LEW | 0.468 |
|
LS v. LEW & SHR |
|
|
FS v. F344 & SHR |
|
|
The amount of nicotine intake or food reward earned by each F1 hybrid strain during the last 3 d was compared to the mean expected from the two parental strains. This value for the mean of the two parental strains predicts F1 behaviors if the genetic determinants were additive. Comparisons that achieved statistical significance (p<0.05) are in bold and italics.
To develop an animal behavioral model capable of elucidating the role of genetics in the vulnerability to smoke cigarettes during adolescence, we focused on nicotine, the principal psychoactive agent in tobacco smoke. We found that nicotine-reinforced operant behavior varies significantly across a panel of 12 isogenic strains of rats in mid-to-late adolescence. These included 6 unique F1 hybrids. These 12 strains displayed a full spectrum of motivated nicotine intake, ranging from an average of 1.4 to 13.0 infusions per 2 h session. This variation in behavior across isogenic strains was largely due to inheritance (0.64). We also found no correlation between motivated nicotine SA and food reward, indicating that genetic control of the motivation to obtain nicotine is distinct from natural rewards.
Genetics plays a major role in the susceptibility to substance abuse and addictive disorders
The inability to control for myriad environmental variables is one of the foremost limitations in detecting the genetic loci that determine the human vulnerability to smoke cigarettes. Rodent models circumvent these limitations, especially when a dedicated team breeds and evaluates all strains in the same facility to avoid inadvertent stressors, such as those due to shipping. Although mice have traditionally been the principal model for genetic studies, establishing nicotine SA in mice is fraught with difficulties. These include not only technical difficulties (e.g., implanting and maintaining a chronically, patent indwelling catheter), but more critically, the challenge of unambiguously attributing the observed behavior directly to i.v. nicotine
The recent availability of rat genomic resources enables genetic studies of rat behavior. For example, the genome sequence of both BN
This study identified a large difference in stable nicotine intake among isogenic strains, with F344 and LEW strains at the extreme ends of the spectrum, respectively. Although the current study used a 2 h limited access model of nicotine SA, the contrasting behavior of the LEW and F344 strains is in agreement with our previous findings obtained from a 23 h model of virtually unlimited access to nicotine SA
In the present studies, F1 hybrids were used to identify additional complexity in the genetic control of nicotine-reinforced operant behavior. Indeed, in four of the F1 crosses, the amount of nicotine intake was different from the expected additive genetic effects of the parental strains (
In summary, we developed a unique panel of isogenic rat strains that effectively model the overall impact of genetics on the vulnerability to acquire nicotine-reinforced behavior during adolescence. The influence of heredity on this process (h2 = 0.64) is similar to that reported for humans. Moreover, in this model, the genetic control of the motivation to obtain nicotine is distinctly different from food reward, indicating the specificity of the underlying genetic mechanisms. Significant gene-gene interactions were found to determine the susceptibility to abuse nicotine in any particular rat strain, as shown by the failure to accurately predict F1 behavior based simply on the inheritance of additive genetic factors from the parental strains. Taken together, these characteristics of the model indicate its strong potential to identify specific genes mediating the human vulnerability to smoke cigarettes, a problem that is exceedingly difficult to resolve by human studies alone.
All procedures were conducted in accordance with the NIH Guidelines concerning the Care and Use of Laboratory Animals, as approved by the Animal Care and Use Committee of the University of Tennessee Health Science Center. Ketoprofen (2 mg/kg, s.c., administered once) was given for post-operative analgesia.
Breeders for six inbred rat strains, including BN, SHR, Dark Agouti (DA), Wistar Kyoto (WKY), Lewis (LEW), and Fisher 344 (F344) were obtained from Harlan Laboratories (Indianapolis, IN). Rats were housed in a 12∶12 h reversed light cycle (lights off at 10∶30 h) with food and water available
For SA, all animals were bred in our animal facility, thereby eliminating the potentially confounding effect of shipping stress on behavior. Only male adolescent offspring were used because several previous studies have found little evidence of sex difference in nicotine SA
Experiments were conducted in operant chambers (Coulbourn Instruments, Whitehall PA, USA) located in sound attenuating enclosures as described previously
Male rats from each strain or F1 cross were randomly assigned to receive either nicotine or food SA. No more than two animals from the same litter were used in a group. For both groups, rats were food-deprived for 24 h on postnatal day (PN) 32, prior to being trained to press a lever to obtain food pellets (45mg, 5TUM Mlab Rodent Tablet, TestDiet). Rats received food equivalent to 10% of their body weight throughout the remainder of the experiment. This initial training terminated when rats received a minimum of 20 pellets on 2 consecutive days, usually realized within 3–4 days.
On PN38, each rat was implanted with a jugular catheter (constructed of PE-60 and silastic tubing), as described previously
Two hour SA sessions were all conducted in operant chambers during the dark phase of the light cycle. This limited access model was preferable to our 23 h access model
Food rewarded behavior was conducted using a similar procedure. Rats responded for a 45 mg food pellet on an FR1 schedule with 60 s timeout during 10 consecutive daily 2 h sessions. The number of parental rats used was 9,8,11,9,7 and 9 for BN, DA, F344, LEW, SHR and WKY inbred strains, respectively, while 10, 10, 9, 9, 8, and 10 rats were used for the FD, FL, FS, LB, LS, and WL F1 offspring, respectively.
Data are presented as mean ± standard error of mean (SEM). The effects of strain on lever press activity and rewards (both nicotine and food) were analyzed using repeated measures ANOVA with day and lever as within subject variables. The correlations between nicotine and food were calculated for the average number of rewards earned and average number of active lever presses emitted by each strain during the last 3 d. All inbred and F1 crosses are isogenic, permitting the calculation of narrow-sense heritability from this dataset. The between-strain variance provides a measure of additive genetic variation (VA), while within-strain variance represents environment variability (VE). An estimate of narrow-sense heritability (i.e. the proportion of total phenotypic variation that is due to the
The authors thank Qingling Wu, Hongxiao Song and Kathy McAllen for their excellent technical assistance.